Since its inception almost 40 years ago, the asset securitization industry has grown to a multitrillion dollar business. The practice of creating multiple tranches from an asset pool has been widely adopted by financial engineers to securitize various assets, including home morgages, automobile loans, credit card receivables, corporate loans, and defaultable bonds.
Tranching makes the securitization very flexible and allows financial intermediaries to design securities with highly heterogeneous characteristics that are “tailor-made” to meet diverse risk/return needs of potential investors. This flexibility naturally leads to the following optimal securitization problem: How can the issuer design securities that optimally make use of investors’ heterogeneity? The goal of this paper is to study this problem.
We capture investors’ heterogeneity through their risk and time preferences as well as their endowments, and we do not limit the potential securities to debt and equity but include general limited liability securities, whose payoffs are backed solely by the given assets. An obvious question is whether reasonable assumptions lead to the optimality of tranching or some other simple securitization procedures.
To better illustrate our approach to the above research question, we describe the overall structure of our model as follows. The issuer possesses some assets, generating a cash flow X with maximal value ¯X at time one. To raise capital at time zero, the issuer designs a basket of N securities, with N equal to the number of types of different investors in the market. Issuing securities is costly, and we allow for both proportional and fixed issuing costs. The security i is a claim to a nonnegative payment Fi = Fi(X), contingent on the realization of X. The securities are backed solely by X, so that the total cash flow generated by all securities never exceeds X.
To focus on the role of clientele effects arising from differences in preferences and endowments, we ignore the effects of asymmetric information and moral hazard and assume that the probability distribution of X is exogenously given and all market participants agree on it. Given the investors’ heterogenous preferences, endowments and reservation utilities, the issuer knows the highest price that each investor is willing to pay for a particular security.
Download
Optimal Securitization with Heterogeneous Investors
